In human-centric manufacturing, collaborative robots (cobots) are increasingly expected to continually acquire knowledge as products and workflows evolve, and learning from human demonstration videos has emerged as a flexible means to this end. Realizing this paradigm requires perception that can be deployed with minimal hardware overhead and that generates executable task knowledge for robotic manipulation. Existing video-based methods rarely satisfy both requirements. Geometric pipelines rely on CAD models or multi-camera rigs, hindering scalable deployment across diverse workpieces, while pre-trained vision-language models require demonstration corpora rarely available in high-mix, low-volume manufacturing. In addition, existing methods recover only low-level geometry or high-level semantic labels in isolation, leaving a gap between geometric perception and executable task knowledge. To address these limitations, we propose Video2Knowledge, a video-based learning framework that takes a single monocular RGB video as input and distills it into temporally consistent task knowledge, covering action sequences, object state transitions, and spatial relations, without requiring additional sensors, manual CAD models, or task-specific retraining. The framework couples a physics-aware 6D object pose corrector with hand-object interaction analysis, and consolidates the results into a structured knowledge representation that maps directly to robot action sequences. Experimental evaluation on the HA-ViD and Assembly101 datasets demonstrates that the physics-aware correction yields improved pose coherence, while the extracted task knowledge faithfully recovers demonstrated action semantics and relative geometric configurations. A real-world case study further validates the transfer of task knowledge from a monocular demonstration video to physical cobot execution, confirming the practical deployability in manufacturing scenarios.
This paper examines how geopolitical fragmentation disrupts carbon pricing effectiveness in China's clean energy markets. Utilizing a TVP-VAR-SV model (2013–2024), it identifies a significant “financial transmission penalty”, where trade conflicts reduce the positive transmission of carbon pricing signals on renewable energy market valuations by 30% to 70%. This paper demonstrates that this attenuation operates through a dual-channel mechanism. Substantively, trade tensions transform the “energy metals” channel from a positive amplifier into a net suppressor, as supply-chain volatility and critical mineral scarcity increase capital expenditures, squeezing policy dividends. Theoretically, it's argued that policy effectiveness is not merely a function of carbon-pricing design but hinges on how financial markets interpret the interaction between climate signals and geopolitical risks through risk premia and investor confidence. The results reveal a structural shift toward “new resource nationalism”, that green technological leadership paradoxically triggers strategic rivalry. This paper concludes and provides policy implications that sustaining decarbonization requires an adaptive framework integrating carbon pricing, mineral reserves, and financial resilience.
Rubber modified asphalt (RMA) fumes constitute a critical precursor reservoir for atmospheric pollution, posing significant risks to ecosystems. However, current knowledge remains limited regarding the multiphase emissions in RMA fumes and its environmental impacts of phase-shifting dynamics. The results demonstrate that RMA emissions undergo temperature- and CR content-dependent phase shifting, which in turn drive distinct secondary pollutant formation and inhalation exposure potential. Specifically, we introduced a quantitative phase-shifting index to characterize multiphase partitioning behavior. By integrating emission intensity and phase shifting with ozone formation potential, secondary organic aerosol formation potential (SOAFP), and respiratory deposition potential, we identified a critical transition temperature, which dominant mechanism of SOAFP transitions from gas-phase dominance toward coupled gas–liquid contributions. Correspondingly, the liquid-phase contribution to SOAFP increases to 44–49%. These findings provide new insights into how reshape fume emission, establishing a mechanistic framework to support the sustainable formulation of RMA and development of high-temperature emission control strategies.
Automating the encoding of building codes into formal representations is a critical step for compliance checking, yet it is often hindered by the linguistic complexity of regulations. This paper develops a two-stage pipeline that utilises Large Language Models (LLMs) to generate LegalRuleML (LRML) through a Structured Intermediate Representation (SIR). Using a dataset derived from the New Zealand Building Code (NZBC), the paper demonstrates that the SIR-based approach outperforms direct translation, with four fine-tuned models achieving F1 scores of 73.29%–77.23%, representing improvements of up to 2.40% over the current state-of-the-art method. Despite these gains, a large gap remains compared to expert-crafted SIRs (80%+). Analysis reveals that LLMs still struggle with logical structural integrity and that standard cross-entropy loss poorly aligns with formalisation goals. These findings highlight the potential of LLM-generated SIRs while underscoring the necessity of task classification and logic-driven verification in automated rule encoding systems.
Ventilated acoustic metamaterials can provide sound attenuation while maintaining airflow, but their inverse design remains challenging because acoustic insulation and ventilation are coupled with high-dimensional, non-unique topology-performance relationships. This study proposes a generative topology inverse design framework for periodic ventilated acoustic metamaterials. A dataset of binary pixelated unit cell topologies is generated in an axisymmetric design domain with a central ventilation channel, and the first phononic bandgap bounds are obtained using Bloch periodic theory. A conditional β-variational autoencoder is then trained to generate unit cell topologies from prescribed conditions including the lower and upper bounds of the first bandgap and the geometric ventilation metric αv. A frozen forward predictor network further enforces consistency with all three prescribed conditions during training, while αv is independently recalculated from each generated binary topology for post-processing verification. Finite element validation shows that the model can generate multiple distinct valid topologies for the same target performance. For four representative cases, finite structure simulations and impedance tube experiments show increased sound transmission loss near the intended frequency regions. The results demonstrate the effectiveness of the proposed framework for one-to-many topology inverse design of ventilated acoustic metamaterials.
